This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug tree-optimization/44423] [4.5/4.6 Regression] Massive performance regression in SSE code due to SRA
- From: "martin at mpa-garching dot mpg dot de" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 14 Jun 2010 12:46:17 -0000
- Subject: [Bug tree-optimization/44423] [4.5/4.6 Regression] Massive performance regression in SSE code due to SRA
- References: <bug-44423-2736@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #16 from martin at mpa-garching dot mpg dot de 2010-06-14 12:46 -------
(In reply to comment #15)
I have found the problem in the meantime ... it's my mistake, sorry about the
noise :(
The problem is that I did not explicitly zero the arrays in main(), so they
apparently contained NaN or similar nastinesses for the small ARRSZ, and
"usual" numbers for large ARRSZ. Of course the processor chokes on the
"unusual" numbers and takes much longer to execute the code.
I'm not sure whether the zeroing should be added for the regression test case
... but since you check for compiler diagnostic and do not try to run the
resulting executable that's probably not necessary.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44423